What question did this study set out to answer?

This research aims to investigate the use of machine learning models to enhance risk prediction for sudden cardiac death (SCD).

January 9, 2026SensorsOpen Access

The random forest model achieved an F2-score of 0.74 and 97.30% recall, indicating strong potential for improving ICD patient selection for SCD risk.

View Full Paper

Ask AI

Bookmark

Key result

Authors

HIHana Ivandic BPBranimir Pervan MPMislav Puljevic

Discussion

Member takes

Overview

Machine learning enhances risk prediction for sudden cardiac death in patients with ICDs, suggesting improved patient selection.

Key Points

This research aims to investigate the use of machine learning models to enhance risk prediction for sudden cardiac death (SCD).
Analyzed tabular clinical data and device-derived variables from patients with ICDs.
Trained various machine learning models including random forest, Naive Bayes, and logistic regression.
Optimized models for the F2-score to focus on high-risk SCD detection.
Conducted post hoc SHAP value analysis for interpretability.
The random forest model achieved the highest F2-score of 0.74 and recall of 97.30%.
The voting classifier achieved an F2-score of 0.71, indicating strong overall discrimination.
Machine learning models demonstrated significant potential in refining ICD patient selection with high recall rates.

Cite This Study

Ivandic et al. (2025) studied this question. The random forest model achieved an F2-score of 0.74 and 97.30% recall, indicating strong potential for improving ICD patient selection for SCD risk.

synapsesocial.com/papers/69609567f6dae357db7c14a9 https://doi.org/10.3390/s26010086

View Full Paper

Ask AI

Bookmark

January 9, 2026SensorsOpen Access

The random forest model achieved an F2-score of 0.74 and 97.30% recall, indicating strong potential for improving ICD patient selection for SCD risk.

View Full Paper

Ask AI

Bookmark

Key result

The random forest model achieved an F2-score of 0.74 and 97.30% recall, indicating strong potential for improving ICD patient selection for SCD risk.

Authors

HIHana Ivandic BPBranimir Pervan MPMislav Puljevic

Discussion

Member takes

Overview

Machine learning enhances risk prediction for sudden cardiac death in patients with ICDs, suggesting improved patient selection.

Key Points

This research aims to investigate the use of machine learning models to enhance risk prediction for sudden cardiac death (SCD).
Analyzed tabular clinical data and device-derived variables from patients with ICDs.
Trained various machine learning models including random forest, Naive Bayes, and logistic regression.
Optimized models for the F2-score to focus on high-risk SCD detection.
Conducted post hoc SHAP value analysis for interpretability.
The random forest model achieved the highest F2-score of 0.74 and recall of 97.30%.
The voting classifier achieved an F2-score of 0.71, indicating strong overall discrimination.
Machine learning models demonstrated significant potential in refining ICD patient selection with high recall rates.

Cite This Study

Ivandic et al. (2025) studied this question. The random forest model achieved an F2-score of 0.74 and 97.30% recall, indicating strong potential for improving ICD patient selection for SCD risk.

synapsesocial.com/papers/69609567f6dae357db7c14a9 https://doi.org/10.3390/s26010086

Also Consider

Synapse has enriched one closely related paper. Consider it for comparative context:

1Age and Outcomes of Primary Prevention Implantable Cardioverter-Defibrillators in Patients With Nonischemic Systolic Heart Failure2017 · 190 citations

View Full Paper

Ask AI

Bookmark